On Tue, 29 Jul 2025 14:38:40 GMT, fabioromano1 <d...@openjdk.org> wrote:
>> Here we have a kind of opposite case: rad can become very imprecise, up to
>> losing all bits from x and become 0.
>> I think this deserves a quick comment.
>
>> Here we have a kind of opposite case: rad can become very imprecise, up to
>> losing all bits from x and become 0. I think this deserves a quick comment.
>
> Later there is `rad = Math.nextUp(rad)`, that ensures `rad > 0.0`.
Sure.
But L.2024-25 read
// Complete the shift to a multiple of n,
// avoiding to lose more bits than necessary.
which is a bit misleading, as it seems there's some hope that some bits can be
preserved, which is not completely true.
The comment can be as succinct as a one liner, so it is worthwhile IMO.
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/24898#discussion_r2240102989
